ValueError: 'video' is both an index level and a column label, which is ambiguous.

[sivashakthi]

I am trying to run the code as is. But master.py is raising an exception when it calls build_features.main(data_set)., at the line around 98

feature_xy = pd.pivot_table(feature_xy, index='video', columns=['bp']).

I tried to work around it by renaming the columns video and bp, but the code raised another exception because the merged features had "non-finite" values.

The data contains non-finite values.

at the line

ref_stats = features[features.category ==0].groupby(['feature_name','age_bracket', 'part'])['Value']
.apply(norm.fit).reset_index()

in the file compute_surprise.py

Any ideas on how to fix this?

Thanks,
Siva